Table 2.

Training and testing performance for the pre-trained networks with fully connected layers

Transfer learning pre-trained model with FC layers	25 Epochs			50 Epochs			75 Epochs			100 Epochs
	Training	Testing		Training	Testing		Training	Testing		Training	Testing
		P based	I based		P based	I based		P based	I based		P based	I based
VGG16	0.91	0.90	0.96	0.90	0.92	0.97	0.92	0.90	0.95	0.93	0.88	0.95
VGG19	0.89	0.93	0.96	0.89	0.88	0.94	0.90	0.92	0.98	0.91	0.93	0.98
ResNet50	0.96	0.87	0.94	0.98	0.84	0.94	0.98	0.81	0.87	0.99	0.81	0.90
MobileNetV2	0.96	0.81	0.94	0.97	0.81	0.95	0.98	0.79	0.90	0.98	0.79	0.88
NASNetMobile	0.93	0.73	0.81	0.96	0.75	0.82	0.97	0.75	0.80	0.97	0.75	0.81